1. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 PART 2

2. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Precise measurements of: m W, m top

3. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Motivation: W mass and top mass are fundamental parameters of the Standard Model:  since G F,  EM, sin  W are known with high precision, precise measurements of m top and m W constrain radiative corrections and Higgs mass (weakly because of logarithmic dependence) So far : W mass measured at LEP2 and Tevatron top mass measured at the Tevatron top mass measured at the Tevatron radiative corrections  r ~ f (m top 2, log m H )  r  3% Fermi constant measured in muon decay Weinberg angle measured at LEP/SLC Electromagnetic constant measured in atomic transitions, e + e - machines, etc.

4. Fabiola Gianotti, Physics at LHC, Pisa, April 2002

7. m W (from LEP2 + Tevatron) = 80.451  0.033 GeV m top (from Tevatron) = 174.3  5.1 GeV m H dependence in SM through radiative corrections Direct measurements light Higgs is favoured

8. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Year 2007:  m W  25 MeV (0.3 ‰) from LEP/Tevatron  m top  2.5 GeV (1.5 %) from Tevatron Can LHC do better ? YES : thanks to large statistics

9. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Measurement of W mass Method used at hadron colliders different from e + e - colliders W  jet jet : cannot be extracted from QCD jet-jet production  cannot be used W  : since  + X, too many undetected neutrinos  cannot be used only W  e and W   decays are used to measure m W at hadron colliders

10. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 W production at LHC : q’ q W Ex. e,   (pp  W + X)  30 nb ~ 300  10 6 events produced ~ 60  10 6 events selected after analysis cuts one year at low L, per experiment ~ 50 times larger statistics than at Tevatron ~ 6000 times larger statistics than WW at LEP

11. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Since not known (only can be measured through E T miss ), measure transverse mass, i.e. invariant mass of in plane perpendicular to the beam :  E T miss

12. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 W  e events (data) from CDF experiment at the Tevatron

13. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 m T W distribution is sensitive to m W m T W distribution expected in ATLAS m T W (GeV) m W = 79.8 GeV m W = 80.3 GeV  fit experimental distributions with Monte Carlo samples with different values of m W  find m W which best fits data

14. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 CDF data : W   transverse mass From fit to transverse mass distribution: m W = 80.465  0.100 GeV

15. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Uncertainties on m W Statistical error negligible  dominated by systematics (mainly Monte Carlo reliability to reproduce real life): detector performance: lepton energy resolution, lepton energy scale, recoil modeling, etc. physics: p T W,  W,  W, structures functions, background, etc. Dominant error (today at Tevatron, also at LHC): knowledge of lepton energy scale of the detector: if lepton energy scale wrong by 1%, then measured m W wrong by 1%  to achieve  m W  20 MeV (~ 0.2‰) need to know lepton scale to  0.2 ‰  most serious experimental challenge Constrained in situ by using mainly Z  decays (1 Hz at low L per ) : e.g. calibrate the electron energy scale in the EM calorimeter requiring m ee = m Z

16. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Calibration of detector energy scale Example : EM calorimeter CALO e - beam E = 100 GeV E measured if E measured = 100.000 GeV  calorimeter is perfectly calibrated if E measured = 99, 101 GeV  energy scale known to 1% to measure m W to ~ 20 MeV need to know energy scale to 0.2 ‰, i.e. if E electron = 100 GeV then 99.98 GeV < E measured < 100.02 GeV  one of most serious experimental challenges

17. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Calibration strategy: detectors equipped with calibration systems which inject known pulses: calorimeter modules calibrated with test beams of known energy  set the energy scale inside LHC detectors: calorimeter sits behind inner detector  electrons lose energy in material of inner detector  need a final calibration “ in situ ” by using physics samples: e.g. Z  e + e - decays 1/sec at low L constrain m ee = m Z reconstructed known to  10 - 5 from LEP cell out in  check that all cells give same response: if not  correct

18. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Expected precision on m W at LHC Source of uncertainty  m W Statistical error << 2 MeV Physics uncertainties ~ 15 MeV (p T W,  W,  W, …) Detector performance < 10 MeV (energy resolution, lepton identification, etc,) Energy scale 15 MeV Total ~ 25 MeV (per experiment, per channel) Combining both channels ( e  ) and both experiments (ATLAS, CMS),  m W  15 MeV should be achieved. However: very difficult measurement

19. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Measurement of m top Top is most intriguing fermion: -- m top  174 GeV  clues about origin of particle masses ? --  top  1.8 GeV  decays before hadronising Discovered in ‘94 at Tevatron  precise measurements of mass, couplings, etc. just started Top mass spectrum from CDF tt  b bjj eventsS+B B -- u c t d s b  m (t-b)  170 GeV  radiative corrections

20. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Top production at LHC: e.g. g g t t t t q q  (pp  + X)  800 pb 10 7 pairs produced in one year at low L ~ 10 2 times more than at Tevatron measure m top,  tt, BR, V tb, single top, rare decays (e.g. t  Zc), resonances, etc. production is the main background to new physics (SUSY, Higgs)

21. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Top decays: t W b BR  100% in SM -- hadronic channel: both W  jj  6 jet final states. BR  50 % but large QCD multijet background. -- leptonic channel: both W   2 jets + 2 + E T miss final states. BR  10 %. Little kinematic constraints to reconstruct mass. -- semileptonic channel: one W  jj, one W   4 jets + 1 + E T miss final states. BR  40 %. If = e,  gold-plated channel for mass measurement at hadron colliders. In all cases two jets are b-jets  displaced vertices in the inner detector

22. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Example from CDF data : tt  Wb Wb  b bjj event e+ Jet 4 (b) (b) W -, m = 79 GeV W+W+ Secondary vertices  (b-hadrons) ~ 1.5 ps  decay at few mm from primary vertex Detected with high-granularity Si detectors (b-tagging)

23. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Selection of  bW bW  b bjj Require: -- two b-tagged jets -- one lepton p T > 20 GeV -- E T miss > 20 GeV -- two more jets W  jj Then require: -- |m jj -m W | < 20 GeV -- combine jj with b-jets. Choose combination which gives highest p T top t  bjj Note : W  jj can be used to calibrate jet energy scale ATLAS

24. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Expected precision on m top at LHC Source of uncertainty  m top Statistical error << 100 MeV Physics uncertainties ~ 1.3 GeV (background, FSR, ISR, fragmentation, etc. ) Jet scale (b-jets, light-quark jets) ~ 0.8 GeV Total ~ 1.5 GeV (per experiment, per channel) Uncertainty dominated by the knowledge of physics and not of detector.

25. Fabiola Gianotti, Physics at LHC, Pisa, April 2002

26. Searches for the Standard Model Higgs boson

27. Fabiola Gianotti, Physics at LHC, Pisa, April 2002

32. Higgs production at LHC gg fusionWW/ZZ fusion associated associated WH, ZH Cross-section for pp  H + X

33. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Higgs decays Decay branching ratios (BR) m H < 120 GeV: H  dominates 130 GeV < m H < 2 m Z : H  WW (*), ZZ (*) dominate m H > 2 m Z : 1/3 H  ZZ 2/3 H  WW important rare decays : H   N. B.:  H ~ m H 3  H ~ MeV (100 GeV) m H ~100 (600) GeV H f ~ m f

34. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Search strategy Fully hadronic final states dominate but cannot be extracted from large QCD background  look for final states with leptons and photons (despite smaller BR). Main channels: Low mass region (m H < 150 GeV): -- H  : BR ~ 100%    20 pb however: huge QCD background (N S /N B < 10 -5 )  can only be used with additional leptons: W H  , H  X  associated production (   1 pb) -- H   : BR ~ 10 -3    50 fb however: clean channel (N S /N B  10 -2 )

35. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Intermediate mass region (120 GeV  m H  2 m Z ): -- H  WW*   -- H  ZZ*  ~ only two channels which can be extracted from background High mass region ( m H > 2 m Z ): -- H  ZZ  gold-plated channel (~ no background) ! -- H  ZZ ,  jet jet -- H  WW  jet jet larger BR  increase rate for m H > 500 GeV Only two examples discussed here : H   H  4 This mass region is disfavoured by EW data (SM internal consistency if Higgs is so heavy ?)

36. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 H   m H  150 GeV H W*     BR  50 fb m H  100 GeV Select events with two photons in the detector with p T ~ 50 GeV Measure energy and direction of each photon Measure invariant mass of photon pair Plot distribution of m   Higgs should appear as a peak at m H Most challenging channel for LHC electromagnetic calorimeters

37. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Main backgrounds:  production: irreducible (i.e. same final state as signal) e.g. :  60 m  ~ 100 GeV q q   g g    jet + jet jet production where one/two jets fake photons: reducible q g   (s) 00 q e.g. : ~ 10 8

38. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 How can one fight these backgrounds ? Reducible  jet, jet-jet: need excellent  /jet separation (in particular  0 separation) to reject jets faking photons R jet  10 3 needed for    80% ATLAS and CMS have calorimeters with good granularity to separate single  from jets or from  0   Simulation of ATLAS calorimeter With this performance : (  jet + jet-jet)  30%   small

39. Fabiola Gianotti, Physics at LHC, Pisa, April 2002

40. Irreducible  : cannot be reduced. But signal can be extracted from background if mass resolution good enough  H < 10 MeV for m H ~ 100 GeV energy resolution of EM calorimeter resolution of the measurement of the  angle 

41. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 ATLAS EM calorimeter: liquid-argon/lead sampling calorimeter longitudinal segmentation  can measure  direction vertex spread ~ 5.6 cm  m  1.3 GeV m H ~100 GeV CMS EM calorimeter: homogeneous crystal calorimeter no longitudinal segmentation  vertex measured using secondary tracks from spectator partons  difficult at high L  often pick up the wrong vertex  m  0.7 GeV m H ~100 GeV   20%   30%

42. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 CMS crystal calorimeter

43. Fabiola Gianotti, Physics at LHC, Pisa, April 2002

44. Expected performance ATLAS : 100 fb -1 ~ 1000 events in the peak m H (GeV) 100 120 150 Significance 4.4 6.5 4.3 ATLAS, 100 fb -1

45. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 CMS : significance is ~ 10% better thanks to better EM calorimeter resolution 100 fb -1

46. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 H  ZZ (*)  4 120  m H < 700 GeV e,  H Z (*) Z e,  mZmZ “Gold-plated” channel for Higgs discovery at LHC Select events with 4 high-p T leptons (  excluded): e + e - e + e -,          e + e -     Require at least one lepton pair consistent with Z mass Plot 4 invariant mass distribution :  Higgs signal should appear as peak in the mass distribution

47. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Backgrounds: -- irreducible : pp  ZZ (*)  4  m (H  4 )  1-1.5 GeV ATLAS, CMS For m H > 300 GeV  H >  m -- reducible (  ~ 100 fb) : t, t W b Both rejected by asking: -- m ~ m Z -- leptons are isolated -- leptons come from interaction vertex ( leptons from B produced at  1 mm from vertex) g g b b Z

48. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Expected performance Significance : 3-25 (depending on mass) for 30 fb -1 Observation possible up to m H  700 GeV For larger masses: --  (pp  H) decreases --  H > 100 GeV H  ZZ*  4 ATLAS, 30 fb -1

49. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 in CMS 20 fb -1 100 fb -1

50. Fabiola Gianotti, Physics at LHC, Pisa, April 2002

51. Summary of Standard Model Higgs Expected significance for one experiment over mass range  1 TeV LHC can discover SM Higgs over full mass region (S > 5) after  2 years of operation in most regions more than one channel is available detector performance (coverage, energy/momentum resolution, particle identification, etc.) crucial in most cases mass can be measured to  1‰ for m H < 600 GeV

52. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 -- SM Higgs boson can be discovered at  5  with 10 fb -1 / experiment (nominally one year at 10 33 cm -2 s -1 ) for m H  130 GeV -- Discovery faster for larger masses -- Whole mass range can be excluded at 95% CL after ~1 month of running at 10 33 cm -2 s -1. However, it will take time to operate, understand, calibrate ATLAS and CMS  Higgs physics will not be done before 2007-2008 given present machine schedule L is per experiment LEP2 55

53. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 For m H ~ 115 GeV Tevatron needs (optimistic analysis): ~ 2 fb -1 for 95% C.L. exclusion  end 2003 ? ~ 5 fb -1 for 3  obervation  end 2004 ? ~ 15 fb -1 for 5  discovery  end 2007 ? Discovery possible up to m H ~120 GeV 95% C.L. exclusion possible up to m H ~185 GeV Tevatron schedule : -- Run 2A : March 2001-end 2003 : ~ 2 fb -1 /expt. -- Run 2B : middle 2004  ? : ~ 15 fb -1 /expt by end 2007 What about Tevatron ?

54. Fabiola Gianotti, Physics at LHC, Pisa, April 2002

55. 2008 Both machines (Tevatron, LHC) could achieve 5  discovery if m H  115 GeV. Who will find it first ? LHC versus TEVATRON Higgs cross-section ~10-100 higher S/B ~ 5 higher Conservative estimates Less conservative predictions (cross-sections, cut analyis, etc.) (e.g. Neural Network analysis) m H =115 GeV 10 fb -1 S/  B  4.7 m H =115 GeV 10 fb -1 S/  B  5.3 4.7  7 using Tevatron approach Will take lot of time to understand Has lot of time to understand Detectors and physics detectors and physics Ready in 2007 ? 15 fb -1 by 2007 ? Need 3 * “ This does not necessarily means that this is the H mass !”

56. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Let’s assume the Higgs is found; what do we do now ? Want to measure the Higgs properties, e.g.  m H can be measured to 0.1% using precise calorimeter and muon systems of ATLAS and CMS mHmH

57. Fabiola Gianotti, Physics at LHC, Pisa, April 2002 Summary of Part 2 Examples of precision physics at LHC: W mass can be measured to ~15 MeV, top mass to ~ 1.5 GeV Standard Model Higgs boson can be discovered over the full mass region up to 1 TeV in ~ 1 year of operation. Excellent detector performance required:  Higgs searches have driven the LHC detector design. Main channels : H  , H  4 If SM Higgs not found before / at LHC, then alternative methods for electroweak symmetry breaking will have to be found End of Part 2